Digestery
Diet and food › Ultra-Processed Food › Studies

Randomized trial · measured other health measures

Trial in 20 adults: more calories eaten and weight gained on ultra-processed diet

Hall and colleagues · Cell metabolism · 2019

Original title: Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain: An Inpatient Randomized Controlled Trial of Ad Libitum Food Intake.

Under review. A person has not yet checked this summary against the paper. It is written from the PubMed abstract. Until a person checks it, search engines are asked not to list this page.

What the study found

Points to harm

People ate 508 ± 106 kcal/day more on the ultra-processed diet and gained 0.9 ± 0.3 kg. On the unprocessed diet they lost 0.9 ± 0.3 kg.

The study at a glance

Type of study
Randomized trial
Who or what was studied
20 people · 20 adults with stable weight, average age 31, living in a hospital research unit
What was tested
Ultra-processed diet vs unprocessed diet, 2 weeks each in turn, eat as much as wanted
What it measured
Several kinds of outcomes
How long
4 weeks
Trial registration
NCT03407053
Who has a stake
Company stake
One author has been paid to speak at events by food firms, advises Kerry Taste and Nutrition, and a group he is in got funding from Abbott, Nestec and Danone.
Journal
Cell metabolism, 2019
Source
PubMed record 31105044. This page is based on the abstract only.

How much this study counts

  • ✓Randomized: people were put into groups by chance, so the groups start out alike.
  • –Small: 20 people. Small trials can miss real effects or find chance ones.
  • –Short: 4 weeks. Longer-term effects are unknown.
  • •Measured several kinds of outcomes.
  • –Company stake: One author has been paid to speak at events by food firms, advises Kerry Taste and Nutrition, and a group he is in got funding from Abbott, Nestec and Danone.
  • •Summarized from the PubMed abstract only. The full paper may hold details the abstract leaves out.

✓ Strength– Limit• Note

These points come from fixed rules about the type of study, its size and length, what it measured, and who has a stake. Every study on Digestery gets the same rules.

Where this study sits

Each dot is one on-topic paper on this question. Steps near the top answer the question more directly.

↑ Stronger evidence

  1. Trials in people: death and chronic disease

    2

    1 no difference, 1 harm

  2. Trials in people: other measures

    3

    1 mixed, 1 no direction, 1 harm

    ↑ This study is the ringed dot

  3. Trials in people: lab tests or safety

    1

    1 mixed

  4. Population studies

    17

    1 mixed, 16 harm

  5. Blood levels and genes

    0

    None among the papers we read.

↓ Weaker evidence

  • Points to benefit
  • Mixed
  • No difference
  • No direction
  • Points to harm

How it fits with the other research

The short answer to “Are ultra-processed foods bad for your health?”

Linked to worse health; cause not proven.

Confidence: low. Based on 38 on-topic papers.

Read the full answer

Few other papers sit on the same step, so there is little to compare it with.

  • Quoted in “What we found”: Two small feeding trials (20 and nine adults) found weight gain on ultra-processed diets: 0.9 ± 0.3 kg over 2 weeks, and 1.1 kg more over 1 week.
  • Part of the finding “Weight gain shows up in studies and short trials”.
  • Marked as a key paper in the research timeline: Hospital trial: ultra-processed diet led to overeating.
  • Open item for a person: Hall trial NCT03407053 appears in Hall 2019, Sciarrillo 2024 and 37276937 (secondary analyses); it is one trial, not three.

Other papers on the same step

See all 38 studies on this question

How this page was made

Label

First read of the abstract
Claude Sonnet
Independent second read
Claude Haiku
Differences settled
The two reads agreed on every checked field. Claude Opus 5.5.
Headline
Claude Sonnet, from the checked study record. A script checked its numbers and wording.
Numbers
Checked against the abstract by a script
“How much this study counts”
Fixed rules, the same for every study
Source check by a person
Not yet. Required before search engines may list this page.

Spot an error? Email a flag with a link to your source, or use the form on the question page.